Week of November 6, 2017

The Conundrum Synthesis Edition

Editor's Note: This week I attended a 2-day AspenEPIC meeting on consumer debt (in the US) and then a day with the Filene Institute on the "Mind-Money Connection." This week's title is inspired by some of Ray Boshara's comments at EPIC about conundrums in understanding consumer debt. But both events once again illustrated the desperate need for more synthesis of ideas and experience between the developing world and the developed world on financial inclusion. Ray also pointed out to me that while I introduced myself when I took over writing the faiV nearly 2 years ago, it's not apparent on a regular basis who the "I" is. So from now on, I'm going to sign these notes each week--Tim Ogden

1. Appropriate Frictions and End-User Behavior: A key theme of the EPIC conversations on debt from my perspective was the importance of differential frictions in access to various kinds of debt. One example: it's much more time consuming to open a home equity line of credit than a credit card account. There are reasons for that of course: we want people to be careful about borrowing against their home, because we fear the consequences for people if they default. But the cost of unsecured credit is so much higher, and various forms of debt are so interlinked, that households can end up in worse straits precisely because we tried to protect them. The true conundrum of appropriate frictions is that the process of determining the best form of credit for a household is in itself a friction that drives consumers toward those willing to provide credit without a care for its impact on the household--a somewhat obtuse but accurate way of describing predatory lenders.
This is one of the lessons from microcredit. Demand for microcredit in most contexts is actually quite low, and rarely did microcredit have much of an impact on local moneylenders. The reason of course being that taking a microloan usually involves a lot of friction, while borrowing from a moneylender is low friction. Those operating in the US will immediately see the exact overlap with payday/auto-title lending vs. working with a community development credit union.
But it's not just a question of the behavior of consumers. Front-line staff also play a role; they are an under-recognized form of end-user that has to be taken into account. Here's some new work by Beisland, D'Espallier and Mersland on "personal mission drift" among credit officers of Ecuadorian MFIs. Now don't look away because this is about microcredit or Ecuador--it's directly applicable to any kind of financial service offered to any kind of customer anywhere. Beisland et al. find that as credit officers gain experience they tend to serve fewer "vulnerable" clients (e.g. smaller loans, young borrowers, disabled borrowers). Why? Because it takes too much time--there are those frictions again. Figuring out how to offer quality products, especially credit, with appropriate frictions for both the borrowers and the credit officer, is a conundrum everywhere.
For further evidence of this, check out the similarities between this piece from Bindu Ananth about conversations with newly banked customers in Indian cities, and this report on "Generational Money Chatter" in the US from Hope Schau and Ignacio Luri (especially from GenXers and Millennials). The common theme I perceive: lots of questions and uncertainties about products and providers, little faith in the "systems," and confusion about where to turn for trustworthy advice.    


2. Frictions, Temptation and Digital Finance: Those of you working in the digital finance world may already be thinking about how digital tools can lower frictions--after all, not only can FinTech tools more quickly and easily gather data from consumers, but they often cut the front-line staff right out of the equation! Take that, friction!
Oh but friction can be useful. This is one of those areas where I'm constantly baffled at the disconnect between the developed and developing worlds. In the developed world, it's generally understood that the goal of payment and digital finance innovation is usually to remove friction specifically for the purpose of getting people to spend more money, more often. Amazon didn't develop and patent one-click ordering out of concern for saving people time (Interesting side note, Amazon's patent on one-click expired last month--exogenous variation klaxon!). The sales pitch that credit card issuers make to merchants has always been that credit cards induce people to spend more.
Here's one of my favorite new pieces of research in a long time: a study of how people in debt management plans handled spending temptation (if that description is too dry to get you to click, try this one: "Target is the Devil!"). The sub-text, and sometimes text, is how hard retailers and some credit providers work to break down the frictions that prevent people from spending.
What's the connection to digital finance, particularly in developing countries. I'll enter there through this piece from Graham Wright based on a debate at the recent MasterCard Foundation Symposium on Financial Inclusion. Graham was asked to make the argument against the hope for digital finance serving poor customers. His list of five reasons why digital finance is "largely irrelevant" in the typical rural village is worth reading at face value. But it's also worth thinking about in terms of how much of digital finance is aimed at removing frictions, how it's failed to remove some of those frictions for poorer customers and what can (or will) happen to poor households when appropriate frictions are removed.

3. Quality Jobs: Another conundrum that deserves more global synthesis is the struggle to create quality jobs for low-income households. Certainly one factor of quality jobs is how much they pay. While there's little doubt that productivity has a big role to play in wages, it's not always clear, particularly since 1973 (the year I was born, coincidence?) how close that link is. Stansbury and Summers have a draft of a paper arguing that the link is still pretty strong. Josh Bivens and Larry Mishel push back, arguing that policies that undermined workers' power led to a divergence of wage growth and productivity growth, and a continuing decline in jobs that pay well enough to be quality jobs.
The stagnant wages for many workers since the 1970s is one of the reasons it's clear to me that it is no longer sufficient to look at having a job as binary. Here's a new review on jobs and recidivism that finds evidence that the quality of job is what matters in helping the formerly incarcerated stay out of prison. Here's a paper from Haltiwanger, Hyatt and McEntarfer on another aspect of job quality--a chance to move up the "job ladder"--and who is getting the opportunity to move up. The surprising news: less-educated workers are more likely to move from a low-productivity firm to a high-productivity one (which should lead to higher wages, but per Mishel et al above, perhaps not).
There's more than one way to take on raising the quality of jobs. Here's work from Akram, Chowdhury and Mobarak on the effect of people moving out of poor areas for better jobs. In short, they are studying a program that subsidizes rural Bangladeshi villagers migrating to cities during the low agricultural season. We already know that raises the wages of the migrants, but it also helps those who don't migrate by tightening the labor market in the villages. Here's where I have to mention that geographic mobility in the US is declining. Meanwhile, take the example of Chicago where segregation continues to shut people out of access to quality jobs, and more. Time for more programs to help Chicagoans migrate, I think, perhaps by reducing some of the frictions.

4. Evidence-Based Policy: Rachel Glennerster examines several cases where evidence led to scaling up programs, asking "When do innovation and evidence change lives?" She doesn't mention the scale-up of the seasonal migration program in Bangladesh mentioned above, but highlights several different models of research and scaling. For those more technically minded, Eva Vivalt has a new version of her paper and an accompanying blog post on research generalizability--how much we can expect a research finding to hold when it is tried elsewhere (or just scaled up). But if you're just looking for examples of research findings that did or didn't hold in differing contexts, take a look at this thread responding to Jess Hoel's appeal for examples to use in her teaching. 
Of course one factor in how generalizable research findings are is the contexts the research is conducted in. David Evans mapped the papers presented at the recent NEUDC conference in three ways (see the comments for the third one). Do you see too much concentration?

5. Neoliberalism: Finally, here's Dani Rodrik on the state of economics and the use of the term neoliberalism. I'm not going to pretend that I can do the piece justice in summary form, so I'll just provide the link and tell you to click on it. Here's the backstory of how the piece ended up in Boston Review.

In case you were too lazy to click on the link to David Evans' mapping of 2017 NEUDC papers, here's the first chart. To get the other two, you're going to have to click. And scroll. Via Development Impact.

In case you were too lazy to click on the link to David Evans' mapping of 2017 NEUDC papers, here's the first chart. To get the other two, you're going to have to click. And scroll. Via Development Impact.

Return to the Weekly faiV